What hardware do you need to run Mistral Small 4 (119B-2603)?

Mistral Small 4 (119B-2603) at Q4_K_M is 74 GB of weights, with a context ceiling of 1024k tokens. Not yet rated: released after our last ratings pass. Despite the name it is a 119B MoE.

Cheapest machine that runs itStrix Halo Framework Desktop, 128GB at $3,449

Fastest of the ones listedMac Studio M5 Ultra, 256GB — 81 tok/s at 32k context

Honest answer on costPays back in 114 years at 500k tokens a day.

How good is it, really?

On the Artificial Analysis Intelligence Index v4.3 it scores 11 (reasoning mode; 9 without), which puts it in the Below every hosted tier band. Fine for simple, well-specified tasks. Noticeably less capable than anything the big labs sell today. Score source. See the whole table.

Summarising — not rated
Translation — not rated
Everyday coding — not rated
Reasoning & maths — not rated
Agentic work — not rated

What it costs either way

Renting the same model costs $0.15 per million input tokens and $0.6 per million output (OpenRouter, cheapest active endpoint, checked 2026-09-09). Buying a machine only beats that if you use it hard enough, for long enough, that the hardware price divides down below the rental bill.

Machines that run it

Machine	Price	Speed at 32k	Pay-back
Strix Halo Framework Desktop, 128GB	$3,449	32 tok/s estimated	Pays back in 114 years	Run the numbers
DGX Spark GB10 Grace Blackwell, 128GB	$4,699	40 tok/s estimated	Pays back in 154 years	Run the numbers
Mac Studio M5 Max, 128GB	$5,099	41 tok/s estimated	Pays back in 166 years	Run the numbers
MacBook Pro M5 Max (16-inch), 128GB	$6,999	41 tok/s estimated	Pays back in 228 years	Run the numbers

One machine per family, cheapest first. Speeds are measured where a public benchmark exists and estimated from memory bandwidth otherwise; the calculator says which for any configuration.

The specifics

Parameters: 119.4B, of which 6B are active per token
Quantisation: Q4_K_M
Weights on disk: 74 GB
KV cache: 0.8 GB at 32k context — Latent attention: the cache holds a 256-wide compressed vector plus 64 RoPE dimensions per layer, 640 bytes a token. The 32 KV heads in the config are not a cache width — reading them that way overstates the memory 25-fold.
Maximum context: 1024k tokens (1M)
Licence: Apache 2.0
Sources: source 1, source 2